Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

WTP for product attributes

Dear Sawtooth Community,

for my master thesis I conducted a Sawtooth CBC aiming on identifying WTP for different product attributes.
In my case, I had 5 attributes:
-ecological production (yes, no)
-regional production (yes, no)
-animal welfare (yes, no)
-fair-trade (yes, no)
-prices (0.79, 0.89, 0.99, 1.09 and 1.19)

Now I want to know the WTP for each of the attributes (yes-level) but I am struggeling on how to calculate this. The only article I could find was this one: https://sawtoothsoftware.com/resources/technical-papers/assessing-the-monetary-value-of-attribute-levels-with-conjoint-analysis-warnings-and-suggestions

are there any more detailed tutorials?
for example, if I understand correctly, I need to specify two products for each of the attributes I want to investigate, one with the attribute and one without. but for example investigating the WTP for eco, which settings do I use for all the other attributes?

I really appreciate your help and any advice!
Best,
Miriam
asked Apr 6 by Miriam
We just released version 9.11 of Lighthouse Studio that can do some automatic willingness to pay estimates for you.  I think we will be releasing some videos on the topic soon, but you can start in the documentation at https://www.sawtoothsoftware.com/help/lighthouse-studio/manual/index.html?understanding-willingness-to-p.html, and make sure to reach the section on Generalized WTP.
Oh, exciting, perfect timing!
I ran the WTP simulation, and now I am wondering if there is a rule of thumb regarding percentage of extrapolated values, and the percentage which did not coverge?

Because my resulting values from the WTP simulation differ greatly from the market simulation as explained by Keith and I am trying to figure out why.

I have extrapolated values between 20 and 40% for the attributes , and between 3 and 30% for the percentage which did not coverge.

Thanks in advance and I am looking forward to the videos on the WTP simulator!
I don't think we have any rules of thumb established yet.  We'll be presenting a paper at our conference later this month which should then get published to the website, but I'm not personally involved with that.

The WTP capabilities will provide different numbers than the single approach that Keith mentioned, because it repeats that approach over and over again with different configurations of the product (and different configurations of the competing products if you do the generalized sampling over scenarios).

1 Answer

0 votes
Miriam,

For calculating WTP you're better off creating a realistic simulation with competitive products.  In it you should include your test product, with the no level and at the lowest price.  Note the share of preference for that test product.

Now change the simulation so that the test product is set to the yes level.  Keep increasing the price until the share of the improved test product again matches the share it had in your first scenario above.  Note the price difference - that's an estimate of your WTP.
answered Apr 6 by Keith Chrzan Platinum Sawtooth Software, Inc. (102,700 points)
Hi Keith,
thanks a lot for your timely answer.
So I understand correctly, I just randomly create some products (how many would you suggest?) and compare these to the "base product"? and designing the base product, I assume it makes a huge difference which other attributes I assign to it?

And for matching the shares, I just have to go with the price levels I set? So for example, the no-level base percentage is 12, and with the yes level I get either 14.6% for one price level or 10.8% for another, which do I take? or does it get any more  specific?

Or can I increase it to in between level prices as well? then I think I might have a problem with the price levels. When I want to assign the exact monetary value, it keeps telling me that levels need to be in ascending or desceneding order. but I unfortunately labeled the levels: -10%, +10%, +20%, +30% & average, so that the order is wrong (average would need to be on value 2 not 5). how can I change the order after conducting the CBC?
The products you create should reflect a realistic mix of competitors.  The WTP you estimate will depend to some extent upon the competition (as you'd expect) so it's best to make them realistic.  

You probably do want to interpolate between your price points.  If you didn't enter them in a monotonic order, then you will have some trouble doing this.  You may want to contact our support department for advice on the steps you'd need to follow to get that to work.
...